Towards Czech Answer Type Analysis

Investor logo

Warning

This publication doesn't include Faculty of Arts. It includes Faculty of Informatics. Official publication website can be found on muni.cz.
Authors

KUŠNIRÁKOVÁ Daša MEDVEĎ Marek

Year of publication 2018
Type Article in Proceedings
Conference Proceedings of the Twelfth Workshop on Recent Advances in Slavonic Natural Languages Processing, RASLAN 2018
MU Faculty or unit

Faculty of Informatics

Citation
Web https://nlp.fi.muni.cz/raslan/2018/paper13-Kusnirakova_Medved.pdf
Keywords question answering; question classification; answer classifica-tion; Czech; Simple Question Answering Database; SQAD
Description In this paper, we introduce two answer type detection systems for Czech language. Based on the input question, the goal of these tools is to recognise the question type and extract an appropriate answer type. Except for the same goal, these systems are completely different. The first one is a rule based system utilising Czech Wordnet for hypernym detection. The second one uses a machine learning approach in form of a neural network. We present architectures of these two systems and offer a detailed evaluation on more than 8,500 question-answer pairs using the SQAD v2.1 benchmark dataset.
Related projects:

You are running an old browser version. We recommend updating your browser to its latest version.